Based on the meteorological elements and environmental air quality monitoring data observed synchronously in the main urban area of Chongqing from 2014 to 2018, the correlation analysis between different atmospheric pollutants and various meteorological factors is carried out, and the meteorological factors affecting the common characteristics of the forecast model are eliminated to identify the meteorological factors significantly affecting air quality. Then combined with the pollutant discharge atmospheric diffusion process and wet sedimentation mechanism, the air quality prediction equation using the mixed mode of power and linear superposition is constructed. The results show that the constructed nonlinear regression equation can accurately reflect the interaction between major atmospheric pollutants and meteorological elements, and the accuracy of the regression model prediction test is satisfactory with a prediction score of 87.6.